Photonic Inverse Design

Automatically Discover Optimal Designs

Take advantage of Inverse Design, a new capability made possible via Lumerical’s Automation API:

Automatically discover optimal geometries for a desired target performance
Two variants available: parametric optimization and topology optimization
Discover non-intuitive geometries that optimize performance, minimize area, and improve manufacturability.

Parametric Optimization

Perform automatic shape-based optimization when the class of the shape is known:

Based on Lumopt, an open-source implementation of the Adjoint Method
Available via the Automation API
Compatible with both FDTD and varFDTD.

Topology Optimization

Topological optimization generates optimal geometries without the need to specify a class of shape
Ideal for non-intuitive problems where an initial “starting point” is not available
Can provide further improvements beyond that of parametric optimization
Continuously monitors manufacturability during optimization

Optimal Designs Faster

Measurements of manufactured designs developed using Lumerical’s Inverse Design have been shown to strongly match simulation results.

Improved Manufacturability

Inverse Design can be configured to meet manufacturing and operating-condition constraints
Simultaneously co-optimize for nominal etching, over-etching, and under-etching for manufacturing tolerance

Accessible via the Automation API

Written in Python, inverse design can interface with across multiple tools
Take advantage of the rich open-source community available via Python

Get Started Today with Application Gallery Examples

Currently-available application-gallery examples make it easy to instantly get started with Inverse Design:

Learn More About Inverse Design